On May 26, 1983, an earthquake, named the Nihonkai-Chubu Earthquake, hit Tohoku area of Japan, and caused large damage of wooden structures in this area mainly because of liquefaction of sandy ground. The magnitude of this earthquake was 7.7, and the largest among the past earthquakes which had occurred on the Japan Sea side. Fortunately, the epicenter of the earthquake was about 100 km west of the sea coast, and the recorded maximum ground acceleration was not more than 200 gal. Except the damage of timber structures caused by liquefaction, very little damage was reported. The Namioka Town Hospital building, at Namioka-cho, Minamitsugaru-gun, Aomori-prefecture, sustained a great deal of destruction by this earthquake. The third and fourth stories of this five story RC building were severely damaged to the extent that emergency repair and strengthening was required. Other building structures near the hospital building, on the other hand, did take almost no damage. In this paper, the outline of overall damage and the damage sustained by the Namioka Town Hospital building are first introduced. And the reasons of the damage are discussed based upon analytical results.

In the analyses, both static and dynamic frame analyses were carried out using two different types of shear deformation characteristics in resisting elements; those are an ordinary origin oriented model and a strength degrading model in which strength degradation is taken into account after reaching their ultimate shear strength.

Concerning dynamic response analyses of this building, simulated earthquake motions in which the maximum acceleration is around 200 gals were applied to in the analyses. These motions were obtained, based on the wave propagation theory, from the bedrock ground motions observed during the earthquake at the Namioka Dam site, which is distant about seven kilometers east from the Namioka Town Hospital building.

The principal conclusions based on dynamic analyses of this hospital building are as follows:

1. Dynamic response of the strength-degrading-type model gives very good agreement with actual damage feautures. On the contrary, the ordinary origin-oriented-type model does not.
2. Two major shocks are believed to have been occurred at approximately 20 seconds interval in the 1983 Nihonkai-Chubu Earthquake. Therefore, the maximum response values for both the first 25 seconds and the last 35 seconds parts of the 60 seconds entire duration time were seperately examined. In this case, the response values of the strength-degrading-type model in the last part greatly increased compared with those in the first part. For instance, story displacement at the third story in the last part is 2.7 centimeters in contrast with only 1.2 centimeters in the first part. It is estimated that damage of the Namioka Town Hospital building would have remarkably decreased in comparison with the actual damage, if the earthquake had been a single shock-type one.
3. A structure which is mainly composed of strength-degrading-type resisting elements shows very unstable behaviour in earthquake response for any input ground motion having considerably large magnitude in acceleration. In case of this hospital building,response for the 225 gals input motion considerably increased compared with the one for the 195 gals input motion. Especially at the third story, the former values greatly increased and reached 1.6 to 2.4 times as much as the latter ones due to only 15% increase of input accelerations in magnitude.

In this report, retrofit design of the Namioka Town Hospital building is also introduced and the effect of the retrofitting are discussed. The retrofittings adopted for the hospital building were mainly ; a. removal and reset of concrete in the shear-failured columns, and b. additional construction of the multi-layered shear walls in the inner frames in the longitudinal direction.

The site investigations of the damaged building and some of the related research works presented herein were carried out by members of the Namioka Town Hospital Building Committee in Building Research Institute.The members of the committee are listed in the following page.